Preventing Repeat Incidents of Family Violence: A Reanalysis of Data From Three Field Tests

• NCJ Number: 200608
• Author(s): Robert C. Davis; Christopher Maxwell
• Date Published: July 2002
• Length: 44 pages
• Publication Series: NIJ Research Report
• Grant Number(s): 2000-WT-VX-0007
• Agencies: NIJ-Sponsored
• Publication Type: Report (Study/Research)

Annotation

This document discusses the reanalysis of inconsistent field tests on repeat victimization of family violence.

Abstract

Domestic violence has the highest repeat rate of all crimes. This is especially true in the first few weeks after an incident is reported to the police. On three separate occasions between 1987 and 1997, three separate field tests were conducted in order to evaluate the basic approaches used in prevention programs in New York City. These field tests were aimed at public housing residents that reported family violence to police. The field-test interventions consisted of a follow-up home visit to households reporting a domestic incident by a police officer and social worker, and a public education program using community meetings, posters, and flyers to educate participants about family violence. The results of all the field tests were inconsistent. Since the composition of the samples varied across studies (two used family violence incidents and the third elder abuse incidents), it could be construed that the prevention programs had different effects with different populations. A series of reanalyses was conducted to try to resolve earlier inconsistencies. The results of the re-analysis of data from three separate field tests of the same interventions unequivocally demonstrate that the interventions cause an increase in reporting of new abusive incidents to authorities and to research interviewers. Those groups assigned to receive home visit or public education interventions reported more abuse than control groups. The fact that the findings were so consistent across the three studies indicates that increased reporting of abuse is not idiosyncratic to one of the samples, but holds across the three different types of samples used in these studies. The results suggest the need for monitoring and strong supervision of programs that intervene in households whose residents have recently reported domestic violence. 9 tables, 3 footnotes, 68 references